fix(pyamber): prevent indefinite worker shutdown hang

[Ma77Ball]

What changes were proposed in this PR?

Hardens the shutdown path of the three StoppableQueueBlockingRunnable threads (MainLoop, NetworkSender, PortStorageWriter) so a stop request no longer depends solely on a single queue wakeup.

• Add a threading.Event stop flag to StoppableQueueBlockingRunnable. stop() now sets the flag and enqueues the existing RUNNABLE_STOP marker. interruptible_get blocks on get(timeout=STOP_POLL_INTERVAL) and treats a queue.Empty timeout as a cue to re-check the flag and exit.
• Plumb an optional timeout through Getable.get, InternalQueue.get, and LinkedBlockingMultiQueue.get (via Condition.wait_for). The default timeout=None preserves the existing blocking behavior exactly, so the tuple data path is untouched and only the stoppable threads opt into polling.

Why?

This is defensive hardening, not a fix for a currently-reproducible hang. As the code stands today, stop() always sets the flag and enqueues the marker through correctly-locked queues, so the wakeup is reliably delivered and no thread parks indefinitely. The value of the change is to decouple "stop was requested" (the flag, checked on a timer) from "the queue delivered the marker" (the notify), so that a future refactor (a new stop path that forgets the marker, or a change to the queue's notify logic) cannot silently reintroduce a shutdown hang.

Performance

• Data/tuple path: no change (timeout=None everywhere except the stop threads).
• Shutdown latency: unchanged. The marker still wakes the thread immediately; the timeout is a fallback not hit in normal operation.
• Idle cost: each stop-eligible thread wakes once per STOP_POLL_INTERVAL (1s) to re-check the flag instead of sleeping indefinitely, a no-op wakeup with negligible CPU cost. The interval can be raised if needed.

Any related issues, documentation, or discussions?

Closes: #5325

How was this PR tested?

• test_stoppable_queue_blocking_thread.py: items reach receive() and stop() ends run(); and setting only the stop flag (no marker) still terminates run(), exercising the timeout recheck path. This last case injects a state the production code does not currently produce, verifying the safety-net mechanism directly.
• Added 3 timeout cases to test_linked_blocking_multi_queue.py (raises Empty on expiry, returns an available item, returns an item arriving mid-wait); the file is 17 passing tests.

Was this PR authored or co-authored using generative AI tooling?

Co-authored with Claude Opus 4.7 in compliance with ASF

[codecov-commenter]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 49.13%. Comparing base (6d604f3) to head (29432ed).
✅ All tests successful. No failed tests found.

Additional details and impacted files

@@             Coverage Diff              @@
##               main    #5326      +/-   ##
============================================
- Coverage     51.15%   49.13%   -2.02%     
+ Complexity     2413     2379      -34     
============================================
  Files          1054     1051       -3     
  Lines         40923    40205     -718     
  Branches       4381     4271     -110     
============================================
- Hits          20933    19756    -1177     
- Misses        18791    19292     +501     
+ Partials       1199     1157      -42     

Flag	Coverage Δ		*Carryforward flag
access-control-service	41.89% <ø> (ø)		Carriedforward from 4cc7539
agent-service	33.76% <ø> (ø)		Carriedforward from 4cc7539
amber	51.56% <ø> (-0.07%)	⬇️	Carriedforward from 4cc7539
computing-unit-managing-service	0.00% <ø> (-1.39%)	⬇️	Carriedforward from 4cc7539
config-service	0.00% <ø> (-54.69%)	⬇️	Carriedforward from 4cc7539
file-service	38.42% <ø> (ø)		Carriedforward from 4cc7539
frontend	40.95% <ø> (-4.81%)	⬇️	Carriedforward from 4cc7539
pyamber	90.72% <100.00%> (+0.03%)	⬆️
python	90.81% <ø> (-0.04%)	⬇️	Carriedforward from 4cc7539
workflow-compiling-service	57.25% <ø> (-1.15%)	⬇️	Carriedforward from 4cc7539

*This pull request uses carry forward flags. Click here to find out more.

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

🚀 New features to boost your workflow:

• 📦 JS Bundle Analysis: Save yourself from yourself by tracking and limiting bundle sizes in JS merges.